Patent Grant
10250958
The present disclosure relates to the field of information technologies, and more specifically, to an optical network-on-chip, an optical router, and a signal transmission method.
In future design of a multi-core processor, because of standstill of a solution for performance and power consumption, on-chip electrical interconnection becomes a bottleneck impeding system development. As silicon-based optoelectronic technologies become mature, optical communication becomes a most promising solution for interconnection of on-chip cores. However, because optical storage cannot be effectively integrated, and an optical logical processing technology is still not mature, communication resources need to be pre-allocated. In an optical network-on-chip (ONoC), an optical circuit exchange mechanism is usually used to implement communication between intellectual property (IP) cores. Before sending data to a destination node (a destination IP core), a source node (a source IP core) needs to reserve a communication resource, and a reserved communication resource is exclusively used by the data sent by the source node to the destination node, that is, another communications node cannot share the resource. Therefore, link utilization of optical circuit exchange is relatively low, and network congestion is relatively severe.
Currently, in an optical network-on-chip in which the optical circuit exchange mechanism is used, Mesh and Torus topologies are widely used. Network diameters (that is, maximum transmission hop counts) of optical networks-on-chip of an N×N scale implemented based on the Mesh and Torus topologies are respectively 2N−2 and N, and in addition, average hop counts of the two networks rapidly increase with expansion of a network scale, and consequently, severe congestion occurs in a process of resource reservation, and further network performance is affected.
Embodiments of the present disclosure provide an optical network-on-chip, an optical router, and a signal transmission method, such that network performance can be improved.
According to a first aspect, an optical network-on-chip is provided, including: N2 intellectual property IP cores, N2/2 gateways, and N2 optical routers, where: N is an even number. The N2 optical routers form two subnets, and every N2/2 optical routers form one subnet. Each gateway in the N2/2 gateways is connected to every two IP cores in the N2 IP cores, where IP cores connected to different gateways are different, and the two IP cores connected to each gateway are in one-to-one correspondences with the two subnets. The N2/2 gateways are in one-to-one correspondences with the N2/2 optical routers in each subnet in the two subnets, where each gateway is connected to an optical router that is in each subnet and that is corresponding to each gateway. A first IP core in the N2 IP cores is configured to send a first signal to a gateway connected to the first IP core. The gateway connected to the first IP core is configured to: determine a second IP core according to the first signal, where the second IP core is a destination IP core of the first signal; and determine a subnet corresponding to the second IP core, and send the first signal to a first optical router in the subnet corresponding to the second IP core, where the first optical router is an optical router connected to the gateway that is connected to the first IP core. The subnet corresponding to the second IP core is configured to: receive the first signal using the first optical router, and route the first signal to a second optical router in the subnet corresponding to the second IP core, where the second optical router is an optical router connected to a gateway that is connected to the second IP core; and send, using the second optical router, the first signal to the gateway connected to the second IP core. The gateway connected to the second IP core is configured to send the first signal to the second IP core.
With reference to the first aspect, in a first possible implementation manner, the gateway connected to the first IP core is configured to determine, according to a correspondence between an IP core and a subnet, the subnet corresponding to the second IP core; and the gateway connected to the second IP core is configured to send the first signal to the second IP core according to the correspondence between an IP core and a subnet.
With reference to the first aspect or the first possible implementation manner of the first aspect, in a second possible implementation manner, the N2 optical routers form the two subnets in the following manner: arranging the N2 optical routers into N rows and N columns, where optical routers located in odd-numbered rows and odd-numbered columns and optical routers located in even-numbered rows and even-numbered columns belong to one subnet, and optical routers located in odd-numbered rows and even-numbered columns and optical routers located in even-numbered rows and odd-numbered columns belong to the other subnet; and in each subnet, each optical router stretches out a pair of parallel waveguides respectively in four directions of upper left, lower left, upper right, and lower right, where the pair of parallel waveguides are connected to waveguides stretched out by another optical router, and if the waveguides arrive at an included angle between subnet edges, the waveguides continue extending in an opposite direction, or if the waveguides arrive at a subnet edge, the waveguides continue extending in a direction perpendicular to the edge.
With reference to the second possible implementation manner of the first aspect, in a third possible implementation manner, N2/2 optical routers in a same subnet in the two subnets have same Z coordinates, and optical routers in different subnets have different Z coordinates; X and Y coordinates of each optical router in each subnet are allocated in the following manner: transforming each subnet to a Torus structure of N rows and N/2 columns, where optical routers in each column are in a longitudinal torus, and optical routers in an ith row and an (i+N/2)th row are in a transverse torus, where i=1,2, . . . , N/2; and successively allocating X and Y coordinates to each optical router in each subnet using an optical router located in an upper left corner as an origin of coordinates, a horizontally rightward direction as an X positive direction, and a vertically downward direction as a Y positive direction, and returning to the original structure according to an inverse process, so as to complete allocation of the coordinates; each IP core in the two IP cores connected to each gateway and the optical router that is connected to each gateway and that is in the subnet corresponding to each IP core have same X, Y, and Z coordinates; the gateway connected to the first IP core is configured to determine, according to Z coordinates of the second IP core, the subnet corresponding to the second IP core; and each optical router in each subnet uses the following routing algorithm:
obtaining current node coordinates (Xc,Yc), and destination node coordinates (Xd,Yd);
if Yc=Yd and Xc=Xd, an output direction is local;
if Yc=Yd and Xc<Xd, an output direction is east;
if Yc=Yd and Xc≥Xd, an output direction is west;
if Yc≠Yd, |Yc−Yd|=N/2, and Xc<Xd, an output direction is west;
if Yc≠Yd, |Yc−Yd|=N/2, and Yc≥Yd, an output direction is east;
if Yc≠Yd, |Yc−Yd|≠N/2, and (Yt−Yc+N)% N>N/2, an output direction is north; or
if Yc≠Yd, |Xc−Yd|≠N/2, and (Yt−Yc+N)% N≤N/2, an output direction is south, where (Xt,Yt) is determined in the following manner:
if Yd<N/2, letting (Xt1,Yt1)=(Xc,Yd), and (Xt2,Yt2)=(Xc,Yd+N/2);
otherwise, letting (Xt1,Yt1)=(Xc,Yd), and (Xt2,Yt2)=(Xc,Yd|N/2); and
calculating Mint1=min(|Yt1|Yc|,N|Yt1−Yc|), and Mint2=min(|Yt2|Yc|,N|Yt2−Yc|);
if Mint1<Mint2, (Xt,Yt)=(Xt1,Tt1); otherwise, (Xt,Yt)=(Xt2,Yt2).
With reference to any one of the first aspect, or the first to the third possible implementation manners of the first aspect, in a fourth possible implementation manner, each optical router in each subnet includes: a first microring resonator, a second microring resonator, a third microring resonator, a fourth microring resonator, a fifth microring resonator, a sixth microring resonator, a seventh microring resonator, an eighth microring resonator, a ninth microring resonator, a tenth microring resonator, an eleventh microring resonator, and a twelfth microring resonator; a first waveguide, a second waveguide, a third waveguide, a fourth waveguide, a fifth waveguide, and a sixth waveguide. One end of the first waveguide is a northern input end, and the other end is a southern output end. One end of the second waveguide is a southern input end, and the other end is a northern output end. One end of the third waveguide is a western input end, and the other end is an eastern output end. One end of the fourth waveguide is an eastern input end, and the other end is a western output end. One end of the fifth waveguide is a local input end. One end of the sixth waveguide is a local output end. A signal that is input at the northern input end arrives at the local output end through the first waveguide, the first microring resonator, and the sixth waveguide. A signal that is input at the southern input end arrives at the local output end through the second waveguide, the fourth microring resonator, and the sixth waveguide. A signal that is input at the western input end arrives at the local output end through the third waveguide, the third microring resonator, and the sixth waveguide. A signal that is input at the eastern input end arrives at the local output end through the fourth waveguide, the second microring resonator, and the sixth waveguide. A signal that is input at the local input end arrives at the eastern output end through the fifth waveguide, the eighth microring resonator, and the third waveguide. The signal that is input at the local input end arrives at the western output end through the fifth waveguide, the fifth microring resonator, and the fourth waveguide. The signal that is input at the local input end arrives at the southern output end through the fifth waveguide, the sixth microring resonator, and the first waveguide. The signal that is input at the local input end arrives at the northern output end through the fifth waveguide, the seventh microring resonator, and the second waveguide. The signal that is input at the northern input end arrives at the western output end through the first waveguide, the ninth microring resonator, and the fourth waveguide. The signal that is input at the northern input end arrives at the eastern output end through the first waveguide, the tenth microring resonator, and the third waveguide. The signal that is input at the southern input end arrives at the western output end through the second waveguide, the twelfth microring resonator, and the fourth waveguide. The signal that is input at the southern input end arrives at the eastern output end through the second waveguide, the eleventh microring resonator, and the third waveguide.
With reference to the second possible implementation manner of the first aspect, in a fifth possible implementation manner, N2/2 optical routers in a same subnet in the two subnets have same Z coordinates, and optical routers in different subnets have different Z coordinates; X and Y coordinates of each optical router in each subnet are allocated in the following manner: transforming each subnet to a Torus structure of N/2 rows and N columns, where optical routers in each row are in a transverse torus, and optical routers in a jt column and a (j+N/2)th column are in a longitudinal torus, where j=1,2, . . . , N/2; and successively allocating X and Y coordinates to each optical router in each subnet using an optical router located in an upper left corner as an origin of coordinates, a horizontally rightward direction as an X positive direction, and a vertically downward direction as a Y positive direction, and returning to the original structure according to an inverse process, so as to complete allocation of the coordinates; each IP core in the two IP cores connected to each gateway and the optical router that is connected to each gateway and that is in the subnet corresponding to each IP core have same X, Y, and Z coordinates; the gateway connected to the first IP core is configured to determine, according to Z coordinates of the second IP core, the subnet corresponding to the second IP core; and each optical router in each subnet uses the following routing algorithm:
obtaining current node coordinates (Xc,Yc), and destination node coordinates (Xd,Yd);
if Xc=Xd and Yc=Yd, an output direction is local;
if Xc=Xd and Yc<Yd, an output direction is south;
if Xc=Xd and Yc≥Yd, an output direction is north;
if Xc≠Xd, |Xc−Xd|=N/2, and Yc<Yd, an output direction is north;
if Xc≠Xd, |Xc−Xd|=N/2, and Yc≥Yd, an output direction is south;
if Xc≠Xd, |Xc−Xd|≠N/2, and (Xt−Xc+N)% N>N/2, an output direction is west; or
if Xc≠Xd, |Xc−Xd|≠N/2, and (Xt−Xc+N)% N≤N/2, an output direction is east, where (Xt,Yt) is determined in the following manner:
if Xd<N/2, letting (Xt1,Yt1)=(Xd,Yc), and (Xt2,Yt2)=(Xd+N/2,Yc);
otherwise, letting (Xt1,Yt1)=(Xd,Yc), and (Xt2,Yt2)=(Xd−N/2,Yc); and
calculating Mint1=min(|Xt1−Xc|,N−|Xt1−Xc|), and Mint2=min(|Xt2|Xc|,N−|Xt2−Xc|);
if Mint1<Mint2, (Xt,Yt)=(Xt1,Tt1); otherwise, (Xt,Yt)=(Xt2,Yt2).
With reference to the first aspect or the first possible implementation manner of the first aspect, in a sixth possible implementation manner, each subnet is a Torus topology.
With reference to the sixth possible implementation manner of the first aspect, in a seventh possible implementation manner, each optical router in each subnet uses an XY routing algorithm or a YX routing algorithm.
With reference to any one of the first aspect, or the first, the second, the fifth to the seventh possible implementation manners of the first aspect, in an eighth possible implementation manner, each optical router in each subnet includes: a first microring resonator, a second microring resonator, a third microring resonator, a fourth microring resonator, a fifth microring resonator, a sixth microring resonator, a seventh microring resonator, and an eighth microring resonator; a first waveguide, a second waveguide, a third waveguide, a fourth waveguide, a fifth waveguide, and a sixth waveguide. One end of the first waveguide is a northern input end, and the other end is a southern output end; one end of the second waveguide is a southern input end, and the other end is a northern output end. One end of the third waveguide is a western input end, and the other end is an eastern output end. One end of the fourth waveguide is an eastern input end, and the other end is a western output end. One end of the fifth waveguide is a local input end. One end of the sixth waveguide is a local output end. A signal that is input at the northern input end arrives at the local output end through the first waveguide, the first microring resonator 801, and the sixth waveguide. A signal that is input at the southern input end arrives at the local output end through the second waveguide, the second microring resonator, and the sixth waveguide. A signal that is input at the western input end arrives at the local output end through the third waveguide, the eighth microring resonator, the first waveguide, the first microring resonator, and the sixth waveguide. A signal that is input at the eastern input end arrives at the local output end through the fourth waveguide, the fifth microring resonator, the second waveguide, the second microring resonator, and the sixth waveguide. A signal that is input at the local input end arrives at the eastern output end through the fifth waveguide, the fourth microring resonator, and the third waveguide. The signal that is input at the local input end arrives at the western output end through the fifth waveguide, the third microring resonator, and the fourth waveguide. The signal that is input at the local input end arrives at the southern output end through the fifth waveguide, the fourth microring resonator, the third waveguide, the eighth microring resonator, and the first waveguide. The signal that is input at the local input end arrives at the northern output end through the fifth waveguide, the third microring resonator, the fourth waveguide, the fifth microring resonator, and the second waveguide. The signal that is input at the eastern input end arrives at the northern output end through the fourth waveguide, the fifth microring resonator, and the second waveguide. The signal that is input at the eastern input end arrives at the southern output end through the fourth waveguide, the sixth microring resonator, and the first waveguide. The signal that is input at the western input end arrives at the northern output end through the third waveguide, the seventh microring resonator, and the second waveguide. The signal that is input at the western input end arrives at the southern output end through the third waveguide, the eighth microring resonator, and the first waveguide.
With reference to any one of the first aspect, or the first, the second, the fifth to the seventh possible implementation manners of the first aspect, in a ninth possible implementation manner, each optical router in each subnet is a Crux optical router.
According to a second aspect, an optical router is provided. The router includes a first microring resonator, a second microring resonator, a third microring resonator, a fourth microring resonator, a fifth microring resonator, a sixth microring resonator, a seventh microring resonator, and an eighth microring resonator. The router also includes a first waveguide, a second waveguide, a third waveguide, a fourth waveguide, a fifth waveguide, and a sixth waveguide. One end of the first waveguide is a northern input end, and the other end is a southern output end. One end of the second waveguide is a southern input end, and the other end is a northern output end. One end of the third waveguide is a western input end, and the other end is an eastern output end. One end of the fourth waveguide is an eastern input end, and the other end is a western output end. One end of the fifth waveguide is a local input end. One end of the sixth waveguide is a local output end. A signal that is input at the northern input end arrives at the local output end through the first waveguide, the first microring resonator, and the sixth waveguide. A signal that is input at the southern input end arrives at the local output end through the second waveguide, the second microring resonator, and the sixth waveguide. A signal that is input at the western input end arrives at the local output end through the third waveguide, the eighth microring resonator, the first waveguide, the first microring resonator, and the sixth waveguide. A signal that is input at the eastern input end arrives at the local output end through the fourth waveguide, the fifth microring resonator, the second waveguide, the second microring resonator, and the sixth waveguide. A signal that is input at the local input end arrives at the eastern output end through the fifth waveguide, the fourth microring resonator, and the third waveguide. The signal that is input at the local input end arrives at the western output end through the fifth waveguide, the third microring resonator, and the fourth waveguide. The signal that is input at the local input end arrives at the southern output end through the fifth waveguide, the fourth microring resonator, the third waveguide, the eighth microring resonator, and the first waveguide. The signal that is input at the local input end arrives at the northern output end through the fifth waveguide, the third microring resonator, the fourth waveguide, the fifth microring resonator, and the second waveguide. The signal that is input at the eastern input end arrives at the northern output end through the fourth waveguide, the fifth microring resonator, and the second waveguide. The signal that is input at the eastern input end arrives at the southern output end through the fourth waveguide, the sixth microring resonator, and the first waveguide. The signal that is input at the western input end arrives at the northern output end through the third waveguide, the seventh microring resonator, and the second waveguide. The signal that is input at the western input end arrives at the southern output end through the third waveguide, the eighth microring resonator, and the first waveguide.
According to a third aspect, an optical router is provided. The router includes a first microring resonator, a second microring resonator, a third microring resonator, a fourth microring resonator, a fifth microring resonator, a sixth microring resonator, a seventh microring resonator, an eighth microring resonator, a ninth microring resonator, a tenth microring resonator, an eleventh microring resonator, and a twelfth microring resonator. The router also includes a first waveguide, a second waveguide, a third waveguide, a fourth waveguide, a fifth waveguide, and a sixth waveguide. One end of the first waveguide is a northern input end, and the other end is a southern output end. One end of the second waveguide is a southern input end, and the other end is a northern output end. One end of the third waveguide is a western input end, and the other end is an eastern output end. One end of the fourth waveguide is an eastern input end and the other end is a western output end. One end of the fifth waveguide is a local input end. One end of the sixth waveguide is a local output end. A signal that is input at the northern input end arrives at the local output end through the first waveguide, the first microring resonator, and the sixth waveguide. A signal that is input at the southern input end arrives at the local output end through the second waveguide, the fourth microring resonator, and the sixth waveguide. A signal that is input at the western input end arrives at the local output end through the third waveguide, the third microring resonator, and the sixth waveguide. A signal that is input at the eastern input end arrives at the local output end through the fourth waveguide, the second microring resonator, and the sixth waveguide. A signal that is input at the local input end arrives at the eastern output end through the fifth waveguide, the eighth microring resonator, and the third waveguide. The signal that is input at the local input end arrives at the western output end through the fifth waveguide, the fifth microring resonator, and the fourth waveguide. The signal that is input at the local input end arrives at the southern output end through the fifth waveguide, the sixth microring resonator, and the first waveguide. The signal that is input at the local input end arrives at the northern output end through the fifth waveguide, the seventh microring resonator, and the second waveguide. The signal that is input at the northern input end arrives at the western output end through the first waveguide, the ninth microring resonator, and the fourth waveguide. The signal that is input at the northern input end arrives at the eastern output end through the first waveguide, the tenth microring resonator, and the third waveguide. The signal that is input at the southern input end arrives at the western output end through the second waveguide, the twelfth microring resonator, and the fourth waveguide. The signal that is input at the southern input end arrives at the eastern output end through the second waveguide, the eleventh microring resonator, and the third waveguide.
According to a fourth aspect, a signal transmission method in an optical network-on-chip is provided. The optical network-on-chip includes: N2 intellectual property IP cores, N2/2 gateways, and N2 optical routers, where N is an even number. The N2 optical routers form two subnets, and every N2/2 optical routers form one subnet. Each gateway in the N2/2 gateways is connected to every two IP cores in the N2 IP cores, where IP cores connected to different gateways are different, and the two IP cores connected to each gateway are in one-to-one correspondences with the two subnets. The N2/2 gateways are in one-to-one correspondences with the N2/2 optical routers in each subnet in the two subnets, where each gateway is connected to an optical router that is in each subnet and that is corresponding to each gateway. The method includes sending, by a first IP core in the N2 IP cores, a first signal to a gateway connected to the first IP core. The method also includes determining, by the gateway connected to the first IP core, a second IP core according to the first signal, where the second IP core is a destination IP core of the first signal; and determining a subnet corresponding to the second IP core, and sending the first signal to a first optical router in the subnet corresponding to the second IP core, where the first optical router is an optical router connected to the gateway that is connected to the first IP core. The method also includes receiving, by the subnet corresponding to the second IP core, the first signal using the first optical router, and routing the first signal to a second optical router in the subnet corresponding to the second IP core, where the second optical router is an optical router connected to a gateway that is connected to the second IP core. The method also includes sending, using the second optical router, the first signal to the gateway connected to the second IP core. The method also includes sending, by the gateway connected to the second IP core, the first signal to the second IP core.
With reference to the fourth aspect, in a first possible implementation manner, the determining a subnet corresponding to the second IP core includes: determining, according to a correspondence between an IP core and a subnet, the subnet corresponding to the second IP core; and sending, by the gateway connected to the second IP core, the first signal to the second IP core includes: sending, by the gateway connected to the second IP core, the first signal to the second IP core according to the correspondence between an IP core and a subnet.
With reference to the fourth aspect or the first possible implementation manner of the fourth aspect, in a second possible implementation manner, the N2 optical routers form the two subnets in the following manner: arranging the N2 optical routers into N rows and N columns, where optical routers located in odd-numbered rows and odd-numbered columns and optical routers located in even-numbered rows and even-numbered columns belong to one subnet, and optical routers located in odd-numbered rows and even-numbered columns and optical routers located in even-numbered rows and odd-numbered columns belong to the other subnet; and in each subnet, each optical router stretches out a pair of parallel waveguides respectively in four directions of upper left, lower left, upper right, and lower right, where the pair of parallel waveguides are connected to waveguides stretched out by another optical router, and if the waveguides arrive at an included angle between subnet edges, the waveguides continue extending in an opposite direction, or if the waveguides arrive at a subnet edge, the waveguides continue extending in a direction perpendicular to the edge.
With reference to the second possible implementation manner of the fourth aspect, in a third possible implementation manner, N2/2 optical routers in a same subnet in the two subnets have same Z coordinates, and optical routers in different subnets have different Z coordinates; X and Y coordinates of each optical router in each subnet are allocated in the following manner: transforming each subnet to a Torus structure of N/2 rows and N columns, where optical routers in each row are in a transverse torus, and optical routers in a jth column and a (j+N/2)th column are in a longitudinal torus, where j=1, 2, . . . , N/2; and successively allocating X and Y coordinates to each optical router in each subnet using an optical router located in an upper left corner as an origin of coordinates, a horizontally rightward direction as an X positive direction, and a vertically downward direction as a Y positive direction, and returning to the original structure according to an inverse process, so as to complete allocation of the coordinates; each IP core in the two IP cores connected to each gateway and the optical router that is connected to each gateway and that is in the subnet corresponding to each IP core have same X, Y, and Z coordinates; the gateway connected to the first IP core determines, according to Z coordinates of the second IP core, the subnet corresponding to the second IP core; and each optical router in each subnet uses the following routing algorithm:
obtaining current node coordinates (Xc,Yc), and destination node coordinates (Xd,Yd);
if Xc=Xd and Yc=Yd, an output direction is local;
if Xc=Xd and Yc<Yd, an output direction is south;
if Xc=Xd and Yc≥Yd, an output direction is north;
if Xc≠Xd, |Xc−Xd−=N/2, and Yc<Yd, an output direction is north;
if Xc≠Xd, |Xc−Xd−=N/2, and Yc≥Yd, an output direction is south;
if Xc≠Xd, |Xc−Xd|≠N/2, and (Xt−Xc+N)% N>N/2, an output direction is west; or
if Xc≠Xd, |Xc−Xd|≠N/2, and (Xt−Xc+N)% N≤N/2, an output direction is east, where (Xt,Yt) is determined in the following manner:
if Xd<N/2, letting (Xt1,Tt1)=(Xd,Yc), and (Xt2,Yt2)=(Xd+N/2,Yc); otherwise, letting (Xt1,Yt1)=(Xd,Yc), and (Xt2,Yt2)=(Xd−N/2,Yc); and
calculating Mint1=min(|Xt1−Xc|,N−|Xt1−Xc|), and Mint2=min(|Xt2−Xc|,N−|Xt2−Xc|); if Mint1<Mint2, (Xt,Yt)=(Xt1,Yt1); otherwise, (Xt,Yt)=(Xt2,Yt2).
With reference to the fourth aspect or the first possible implementation manner of the fourth aspect, in a fourth possible implementation manner, each subnet is a Torus topology.
With reference to the fourth possible implementation manner of the fourth aspect, in a fifth possible implementation manner, each optical router in each subnet uses an XY routing algorithm or a YX routing algorithm.
With reference to the second possible implementation manner of the fourth aspect, in a sixth possible implementation manner, N2/2 optical routers in a same subnet in the two subnets have same Z coordinates, and optical routers in different subnets have different Z coordinates; X and Y coordinates of each optical router in each subnet are allocated in the following manner: transforming each subnet to a Torus structure of N rows and N/2 columns, where optical routers in each column are in a longitudinal torus, and optical routers in an ith row and an (i+N/2)th row are in a transverse torus, where i=1,2, . . . , N/2; and successively allocating X and Y coordinates to each optical router in each subnet using an optical router located in an upper left corner as an origin of coordinates, a horizontally rightward direction as an X positive direction, and a vertically downward direction as a Y positive direction, and returning to the original structure according to an inverse process, so as to complete allocation of the coordinates; each IP core in the two IP cores connected to each gateway and the optical router that is connected to each gateway and that is in the subnet corresponding to each IP core have same X, Y, and Z coordinates; the gateway connected to the first IP core determines, according to Z coordinates of the second IP core, the subnet corresponding to the second IP core; and each optical router in each subnet uses the following routing algorithm:
obtaining current node coordinates (Xc,Yc), and destination node coordinates (Xd,Yd);
if Yc=Yd and Xc=Xd, an output direction is local;
if Yc=Yd and Xc<Xd, an output direction is east;
if Yc=Yd and XnXd, an output direction is west;
if Yc*Yd, |Yc−Yd|=N/2, and Xc<Xd, an output direction is west;
if Yc*Yd, |Yc−Yd|=N/2, and Yc≥Yd, an output direction is east;
if Yc*Yd, |Yc−Yd|*N/2, and (Yt−Yc+N)% N>N/2, an output direction is north; or
if Yc*Yd, IXc−Yd|*N/2, and (Yt−Yc+N)% N≤N/2, an output direction is south, where (Xt,Yt) is determined in the following manner:
if Yd<N/2, letting (Xt1,Yt1)=(Xc,Yd), and (Xt2,Yt2)=(Xc,Yd+N/2);
otherwise, letting (Xt1,Yt1)=(Xc,Yd), and (Xt2,Yt2)=(Xc,Yd-N/2); and
calculating Mint1=min(|Yt1−Yc|,N−|Yt1−Yc|), and Mint2=min(|Yt2−Yc|,N−|Yt2−Yc|);
if Mint1<Mint2, (Xt,Yt)=(Xt1,Tt1); otherwise, (Xt,Yt)=(Xt2,Yt2).
Based on the foregoing technical solutions, in the embodiments of the present disclosure, an optical network-on-chip is divided into two subnets, and every two IP cores share one gateway, and the gateway may be connected to different networks according to a requirement, which reduces a network diameter, reduces communication resources required for each communication task, decreases network congestion, and may not only improve network transmission bandwidth, but also reserve a characteristic of efficient partial transmission of data, such that network performance can be improved.
To describe the technical solutions in the embodiments of the present disclosure more clearly, the following briefly describes the accompanying drawings required for describing the embodiments of the present disclosure. Apparently, the accompanying drawings in the following description show merely some embodiments of the present disclosure, and a person of ordinary skill in the art may still derive other drawings from these accompanying drawings without creative efforts.
The following clearly describes the technical solutions in the embodiments of the present disclosure with reference to the accompanying drawings in the embodiments of the present disclosure. Apparently, the described embodiments are a part rather than all of the embodiments of the present disclosure. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present disclosure without creative efforts shall fall within the protection scope of the present disclosure.
As shown in
The N2 optical routers 103 form two subnets, and every N2/2 optical routers form one subnet, that is, a subnet 1 and a subnet 2 in
Each gateway in the N2/2 gateways 102 is connected to every two IP cores in the N2 IP cores 101, where IP cores connected to different gateways are different, and the two IP cores connected to each gateway are in one-to-one correspondences with the two subnets.
That is, every two IP cores are one cluster, and N2 IP cores 101 form N2/2 clusters. Communication may be performed in each cluster in an electrical interconnection manner, and each cluster is connected to the subnet 1 and the subnet 2 using a gateway. Connection between each cluster and the subnet 1 and the subnet 2 is optical connection.
The N2/2 gateways 102 are in one-to-one correspondences with the N2/2 optical routers 103 in each subnet in the two subnets, where each gateway is connected to an optical router that is in each subnet and that is corresponding to each gateway. That is, each gateway is connected to two optical routers, and the two optical routers are respectively located in different subnets.
Each IP core can receive a signal from only a subnet corresponding to the IP core. That is, each gateway may send a signal sent by an optical router that is connected to the gateway and that is in each subnet to an IP core corresponding to the subnet.
Each gateway may send a signal of any IP core in two IP cores connected to the gateway to a subnet corresponding to a destination IP core of the signal, that is, sending the signal to an optical router that is connected to the gateway and that is in the subnet corresponding to the destination IP core of the signal.
After the signal is sent to each subnet, each subnet may route the signal, and finally send the signal to a gateway connected to the destination IP core of the signal, and the gateway sends the signal to the destination IP core.
The first IP core 201 sends the first signal to a gateway 203 connected to the first IP core.
The gateway 203 connected to the first IP core 201 determines the second IP core 202 according to the first signal, and determines a subnet corresponding to the second IP core 202. Optionally, the gateway 203 may determine, according to a correspondence between an IP core and a subnet, the subnet corresponding to the second IP core 202, and the correspondence may be preset. For example, the subnet corresponding to the second IP core 202 is a subnet 2. The gateway 203 sends the first signal to a first optical router 204 in the subnet 2 corresponding to the second IP core 202, and the first optical router 204 is an optical router connected to the gateway 203 in the subnet 2.
The subnet 2 corresponding to the second IP core receives the first signal using the first optical router 204, and routes the first signal to a second optical router 205. The second optical router 205 is an optical router connected to a gateway 206 that is connected to the second IP core 202 in the subnet 2, and the second optical router 205 sends the first signal to the gateway 206 connected to the second IP core.
The gateway 206 connected to the second IP core 202 sends the first signal to the second IP core 202. Optionally, the gateway 206 connected to the second IP core 202 may send the first signal to the second IP core 202 according to the correspondence between an IP core and a subnet.
It can be learned from the foregoing that the signal from the first IP core 201 to the second IP core 202 is routed in only one subnet (the subnet 2), and a network hop count is relatively small. In other words, compared with a network formed by N2 optical routers, a subnet formed by N2/2 optical routers has a smaller network diameter.
Therefore, in this embodiment of the present disclosure, an optical network-on-chip is divided into two subnets, and every two IP cores share one gateway, and the gateway may be connected to different networks according to a requirement, which reduces a network diameter, reduces communication resources required for each communication task, decreases network congestion, and may not only improve network transmission bandwidth, but also reserve a characteristic of efficient partial transmission of data, such that network performance can be improved.
Optionally, in an embodiment of the present disclosure, each subnet may be a Torus topology.
Optionally, each optical router in each subnet may use an XY routing algorithm or a YX routing algorithm.
In the XY routing algorithm, an address of each node is (x,y), and a message is first routed along an X dimension, and then routed along a Y dimension. In the YX routing algorithm, an address of each node is (x,y), and a message is first routed along a Y dimension, and then routed along an X dimension.
Particularly, if a source and a destination are respectively (sx,sy) and (dx,dy), a routing message moves |dx−sx| steps on the X dimension, and then moves |dy−sy| steps on the Y dimension.
Both the XY routing algorithm and the YX routing algorithm belong to the prior art, and therefore are not limited in none of the embodiments of the present disclosure.
It should be understood that each subnet may also be another topology, for example, a Mesh topology, which is not limited in this embodiment of the present disclosure.
Optionally, in an embodiment of the present disclosure, two subnets may be formed in the following manner. As shown in
As shown in
It should be understood that
Optionally, in an embodiment of the present disclosure, coordinates may be allocated to the optical routers and the IP cores that are in the optical network-on-chip loo.
N2/2 optical routers in a same subnet in the two subnets have same Z coordinates, and optical routers in different subnets have different Z coordinates. For example, for an optical router in the subnet 1, Z=0, and for an optical router in the subnet 2, Z=1.
X and Y coordinates of each optical router in each subnet are allocated in the following manner: transforming each subnet to a Torus structure of N/2 rows and N columns, where optical routers in each row are in a transverse torus, and optical routers in a jth column and a (j+N/2)th column are in a longitudinal torus, where j=1, 2, . . . , N/2; and successively allocating X and Y coordinates to each optical router in each subnet using an optical router located in an upper left corner as an origin of coordinates, a horizontally rightward direction as an X positive direction, and a vertically downward direction as a Y positive direction, and returning to the original structure according to an inverse process, so as to complete allocation of the coordinates.
The subnet 1 is used as an example. A structure in
Each IP core in two IP cores connected to each gateway and an optical router that is connected to each gateway and that is in the subnet corresponding to each IP core have same X, Y, and Z coordinates. In other words, for the two IP cores and the two optical routers that are connected to each gateway, coordinates of the two IP cores are different, but are respectively the same as coordinates of the two optical routers. An IP core can receive a signal that is output only by an optical router having the same coordinates as the IP core.
In this case, a gateway connected to a first IP core may determine, according to Z coordinates of a second IP core, a subnet corresponding to the second IP core.
Each optical router in each subnet may use the following routing algorithm: obtaining current node coordinates (Xc,Yc), and destination node coordinates (Xd,Yd);
if Xc=Xd and Yc=Yd, an output direction is local;
if Xc=Xd and Yc<Yd, an output direction is south;
if Xc=Xd and Yc≥Yd, an output direction is north;
if Xc≠Xd, |Xc−Xd|=N/2, and Yc<Yd, an output direction is north;
if Xc≠Xd, |Xc−Xd|=N/2, and Yc≥Yd, an output direction is south;
if Xc≠Xd, |Xc−Xd|≠N/2, and (Xt−Xc+N)% N>N/2, an output direction is west; or
if Xc≠Xd, |Xc−Xd|≠N/2, and (Xt−Xc+N)% N≤N/2, an output direction is east, where (Xt,Yt) is determined in the following manner:
if Xd<N/2, letting (Xt1,Yt1)=(Xd,Yc), and (Xt2,Yt2)=(Xd+N/2,Yc);
otherwise, letting (Xt1,Yt1)=(Xd,Yc), and (Xt2,Yt2)=(Xd−N/2,Yc); and
calculating Mint1=min(|Xt1−Xc|,N−|Xt1−Xc|), and Mint2 =min(|Xt2−Xc|,N−|Xt2−Xc|);
if Mint1<Mint2, (Xt,Yt)=(Xt1,Yt1); otherwise, (Xt,Yt)=(Xt2,Yt2).
701. Obtain current node coordinates (Xc,Yc), and destination node coordinates (Xd,Yd).
702. Determine whether (Xc=Xd) or whether |Xc−Xd|=N/2; if yes, go to 703, or if no, go to 712.
703. Determine whether Xc=Xd; if yes, go to 704, or if no, go to 709.
704. Determine whether Yc=Yd; if yes, go to 705, or if no, go to 706.
705. An output direction is local.
706. Determine whether Yc<Yd; if yes, go to 707, or if no, go to 708.
707. An output direction is south.
708. An output direction is north.
709. Determine whether Yc<Yd; if yes, go to 710, or if no, go to 711.
710. An output direction is north.
711. An output direction is south.
712. If Xd<N/2, let (Xt1,Yt1)=(Xd,Yc), and (Xt2,Yt2)=(Xd+N/2,Yc); otherwise, let (Xt1,Yt1)=(Xd,Yc), and (Xt2,Yt2)=(Xd−N/2,Yc).
713. Calculate Mint1=min(|Xt1−Xc|,N−|Xt1−Xc|), and Mint2=min(|Xt2−Xc|,N−|Xt2−Xc|).
714. If Mint1<Mint2, let (Xt,Yt)=(Xt1,Yt1); otherwise, let (Xt,Yt)=(Xt2,Yt2).
715. Determine whether (Xt−Xc+N)% N>N/2; if yes, go to 716, or if no, go to 717, where % represents a remainder operation.
716. An output direction is west.
717. An output direction is east.
According to an interconnection structure of optical routers inside a subnet of an optical network-on-chip and a used routing algorithm in this embodiment of the present disclosure, a network diameter is only N/2, which further reduces communication resources required for each communication task, and decreases network congestion.
Optionally, in an embodiment of the present disclosure, each optical router in each subnet may be the following optical router 800 provided in this embodiment of the present disclosure.
One end of the first waveguide 810 is a northern input end 811, and the other end is a southern output end 812.
One end of the second waveguide 820 is a southern input end 821, and the other end is a northern output end 822.
One end of the third waveguide 830 is a western input end 831, and the other end is an eastern output end 832.
One end of the fourth waveguide 840 is an eastern input end 841, and the other end is a western output end 842.
One end of the fifth waveguide 850 is a local input end 851.
One end of the sixth waveguide 860 is a local output end 861.
A signal that is input at the northern input end 811 arrives at the local output end 861 through the first waveguide 810, the first microring resonator 801, and the sixth waveguide 860.
A signal that is input at the southern input end 821 arrives at the local output end 861 through the second waveguide 820, the second microring resonator 802, and the sixth waveguide 860.
A signal that is input at the western input end 831 arrives at the local output end 861 through the third waveguide 830, the eighth microring resonator 808, the first waveguide 810, the first microring resonator 801, and the sixth waveguide 860.
A signal that is input at the eastern input end 841 arrives at the local output end 861 through the fourth waveguide 840, the fifth microring resonator 805, the second waveguide 820, the second microring resonator 802, and the sixth waveguide 860.
A signal that is input at the local input end 851 arrives at the eastern output end 832 through the fifth waveguide 850, the fourth microring resonator 804, and the third waveguide 830.
The signal that is input at the local input end 851 arrives at the western output end 842 through the fifth waveguide 850, the third microring resonator 803, and the fourth waveguide 840.
The signal that is input at the local input end 851 arrives at the southern output end 812 through the fifth waveguide 850, the fourth microring resonator 804, the third waveguide 830, the eighth microring resonator 808, and the first waveguide 810.
The signal that is input at the local input end 851 arrives at the northern output end 822 through the fifth waveguide 850, the third microring resonator 803, the fourth waveguide 840, the fifth microring resonator 805, and the second waveguide 820.
The signal that is input at the eastern input end 841 arrives at the northern output end 822 through the fourth waveguide 840, the fifth microring resonator 805, and the second waveguide 820.
The signal that is input at the eastern input end 841 arrives at the southern output end 812 through the fourth waveguide 840, the sixth microring resonator 806, and the first waveguide 810.
The signal that is input at the western input end 831 arrives at the northern output end 822 through the third waveguide 830, the seventh microring resonator 807, and the second waveguide 820.
The signal that is input at the western input end 831 arrives at the southern output end 812 through the third waveguide 830, the eighth microring resonator 808, and the first waveguide 810.
According to the optical router in this embodiment of the present disclosure, a quantity of used microring resonators is reduced, and utilization of each microring resonator is improved.
It should be understood that in this embodiment of the present disclosure, each optical router in each subnet may also be a Crux optical router, which is not limited in this embodiment of the present disclosure.
Embodiments of the present disclosure further provide a manner of allocating coordinates to optical routers and IP cores that are in an optical network-on-chip 100.
Specifically, N2/2 optical routers in a same subnet in two subnets have same Z coordinates, and optical routers in different subnets have different Z coordinates. For example, for an optical router in the subnet 1, Z=0, and for an optical router in the subnet 2, Z=1.
X and Y coordinates of each optical router in each subnet are allocated in the following manner: transforming each subnet to a Torus structure of N rows and N/2 columns, where optical routers in each column are in a longitudinal torus, and optical routers in an ith row and an (i+N/2)th row are in a transverse torus, where i=1,2, . . . , N/2; and successively allocating X and Y coordinates to each optical router in each subnet using an optical router located in an upper left corner as an origin of coordinates, a horizontally rightward direction as an X positive direction, and a vertically downward direction as a Y positive direction, and returning to the original structure according to an inverse process, so as to complete allocation of the coordinates.
The subnet 1 is used as an example. A structure in
Each IP core in two IP cores connected to each gateway and an optical router that is connected to each gateway and that is in a subnet corresponding to each IP core have same X, Y, and Z coordinates. In other words, for the two IP cores and the two optical routers that are connected to each gateway, coordinates of the two IP cores are different, but are respectively the same as coordinates of the two optical routers. An IP core can receive a signal that is output only by an optical router having the same coordinates as the IP core.
In this case, a gateway connected to a first IP core may determine, according to Z coordinates of a second IP core, a subnet corresponding to the second IP core.
In this embodiment, each optical router in each subnet may use the following routing algorithm:
obtaining current node coordinates (Xc,Yc), and destination node coordinates (Xd,Yd);
if Yc=Yd and Xc=Xd, an output direction is local;
if Yc=Yd and Xc<Xd, an output direction is east;
if Yc=Yd and XnXd, an output direction is west;
if Yc≠Yd, |Yc−Yd|=N/2, and Xc<Xd, an output direction is west;
if Yc≠Yd, |Yc−Yd|=N/2, and Yc≥Yd, an output direction is east;
if Yc≠Yd, |Yc−Yd|≠N/2, and (Yt−Yc+N)% N>N/2, an output direction is north; or
if Yc≠Yd, |Xc−Yd|≠N/2, and (Yt−Yc+N)% N≤N/2, an output direction is south, where (Xt,Yt) is determined in the following manner:
if Yd<N/2, letting (Xt1,Yt1)=(Xc,Yd), and (Xt2,Yt2)=(Xc,Yd+N/2);
otherwise, letting (Xt1,Yt1)=(Xc,Yd), and (Xt2,Yt2)=(Xc,Yd−N/2); and
calculating Mint1=min(|Yt1−Yc|,N−|Yt1−Yc|), and Mint2=min(|Yt2−Yc|,N−|Yt2−Yc|);
if Mint1<Mint2, (Xt,Yt)=(Xt1,Yt1); otherwise, (Xt,Yt)=(Xt2,Yt2).
701
b. Obtain current node coordinates (Xc,Yc), and destination node coordinates (Xd,Yd).
702
b. Determine whether (Yc=Yd) or |Yc−Yd|=N/2; if yes, go to 703b, or if no, go to 712b.
703
b. Determine whether Yc=Yd; if yes, go to 704b, or if no, go to 709b.
704
b. Determine whether Xc=Xd; if yes, go to 705b, or if no, go to 706b.
705
b. An output direction is local.
706
b. Determine whether Xc<Xd; if yes, go to 707b, or if no, go to 708b.
707
b. An output direction is east.
708
b. An output direction is west.
709
b. Determine whether Xc<Xd; if yes, go to 710b, or if no, go to 711b.
710
b. An output direction is west.
711
b. An output direction is east.
712. If Yd<N/2, let (Xt1,Yt1)=(Xc,Yd), and (Xt2,Yt2)=(Xc,Yd+N/2), otherwise, let (Xt1,Yt1)=(Xc,Yd), and (Xt2,Yt2)=(Xc,Yd−N/2).
713
b. Calculate Mint1=min(|Yt1−Yc|,N−|Yt1−Yc|), and Mint2=min(|Yt2−Yc|,N−|Yt2−Yc|).
714
b. If Mint1<Mint2, let (Xt,Yt)=(Xt1,Yt1); otherwise, let (Xt,Yt)=(Xt2,Yt2).
715
b. Determine whether (Yt−Yc+N)% N>N/2; if yes, go to 716b, or if no, go to 717b, where % represents a remainder operation.
716
b. An output direction is north.
717
b. An output direction is south.
Optionally, in this embodiment, each optical router in each subnet may be the following optical router woo provided in an embodiment of the present disclosure.
One end of the first waveguide 1010 is a northern input end 1011, and the other end is a southern output end 1012.
One end of the second waveguide 1020 is a southern input end 1021, and the other end is a northern output end 1022.
One end of the third waveguide 1030 is a western input end 1031, and the other end is an eastern output end 1032.
One end of the fourth waveguide 1040 is an eastern input end 1041, and the other end is a western output end 1042.
One end of the fifth waveguide 1050 is a local input end 1051.
One end of the sixth waveguide 1060 is a local output end 1061.
A signal that is input at the northern input end 1011 arrives at the local output end 1061 through the first waveguide 1010, the first microring resonator 1001, and the sixth waveguide 1060.
A signal that is input at the southern input end 1021 arrives at the local output end 1061 through the second waveguide 1020, the fourth microring resonator 1004, and the sixth waveguide 1060.
A signal that is input at the western input end 1031 arrives at the local output end 1061 through the third waveguide 1030, the third microring resonator 1003, and the sixth waveguide 1060.
A signal that is input at the eastern input end 1041 arrives at the local output end 1061 through the fourth waveguide 1040, the second microring resonator 1002, and the sixth waveguide 1060.
A signal that is input at the local input end 1051 arrives at the eastern output end 1032 through the fifth waveguide 1050, the eighth microring resonator 1008, and the third waveguide 1030.
The signal that is input at the local input end 1051 arrives at the western output end 1042 through the fifth waveguide 1050, the fifth microring resonator 1005, and the fourth waveguide 1040.
The signal that is input at the local input end 1051 arrives at the southern output end 1012 through the fifth waveguide 1050, the sixth microring resonator 1006, and the first waveguide 1010.
The signal that is input at the local input end 1051 arrives at the northern output end 1022 through the fifth waveguide 1050, the seventh microring resonator 1007, and the second waveguide 1020.
The signal that is input at the northern input end 1011 arrives at the western output end 1042 through the first waveguide 1010, the ninth microring resonator 1009, and the fourth waveguide 1040.
The signal that is input at the northern input end 1011 arrives at the eastern output end 1032 through the first waveguide 1010, the tenth microring resonator 1110, and the third waveguide 1030.
The signal that is input at the southern input end 1021 arrives at the western output end 1042 through the second waveguide 1020, the twelfth microring resonator 1112, and the fourth waveguide 1040.
The signal that is input at the southern input end 1021 arrives at the eastern output end 1032 through the second waveguide 1020, the eleventh microring resonator 1111, and the third waveguide 1030.
As shown in
S1210. A first IP core in N2 IP cores sends a first signal to a gateway connected to the first IP core.
S1220. The gateway connected to the first IP core determines a second IP core according to the first signal, where the second IP core is a destination IP core of the first signal; and determines a subnet corresponding to the second IP core, and sends the first signal to a first optical router in the subnet corresponding to the second IP core, where the first optical router is an optical router connected to the gateway that is connected to the first IP core.
S1230. The subnet corresponding to the second IP core receives the first signal using the first optical router, and routes the first signal to a second optical router in the subnet corresponding to the second IP core, where the second optical router is an optical router connected to a gateway that is connected to the second IP core; and the second optical router sends the first signal to the gateway connected to the second IP core.
S1240. The gateway connected to the second IP core sends the first signal to the second IP core.
Optionally, in an embodiment of the present disclosure, the gateway connected to the first IP core determines, according to a correspondence between an IP core and a subnet, the subnet corresponding to the second IP core.
The gateway connected to the second IP core sends the first signal to the second IP core according to the correspondence between an IP core and a subnet.
Corresponding processes of the signal transmission method 1200 according to this embodiment of the present disclosure may be respectively implemented by all parts of the optical network-on-chip 100 in the foregoing embodiments of the present disclosure. For detailed descriptions, refer to the foregoing embodiments, and for brevity, details are not described herein again.
According to the signal transmission method in this embodiment of the present disclosure, communication resources required for each communication task are reduced, and network congestion is decreased, which may not only improve network transmission bandwidth, but also reserve a characteristic of efficient partial transmission of data, such that network performance can be improved.
It should be understood that specific examples in the present disclosure are intended merely to help a person skilled in the art better understand the embodiments of the present disclosure, but are not intended to limit the scope of the embodiments of the present disclosure.
It should also be understood that sequence numbers of the foregoing processes do not mean execution sequences in various embodiments of the present disclosure. The execution sequences of the processes should be determined according to functions and internal logic of the processes, and should not be construed as any limitation on the implementation processes of the embodiments of the present disclosure.
A person of ordinary skill in the art may be aware that, in combination with the examples described in the embodiments disclosed in this specification, units and algorithm steps may be implemented by electronic hardware, computer software, or a combination thereof. To clearly describe the interchangeability between the hardware and the software, the foregoing has generally described compositions and steps of each example according to functions. Whether the functions are performed by hardware or software depends on particular applications and design constraint conditions of the technical solutions. A person skilled in the art may use different methods to implement the described functions for each particular application, but it should not be considered that the implementation goes beyond the scope of the present disclosure.
It may be clearly understood by a person skilled in the art that, for the purpose of convenient and brief description, for a specific process of the method described above, reference may be made to a corresponding description in the foregoing apparatus embodiments, and details are not described herein.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus, and method may be implemented in other manners. For example, the described apparatus embodiment is merely exemplary. For example, the unit division is merely logical function division and may be other division in actual implementation. For example, a plurality of units or components may be combined or integrated into another system, or some features may be ignored or not performed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections may be implemented through some interfaces. The indirect couplings or communication connections between the apparatuses or units may be implemented in electronic, mechanical, or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on a plurality of network units. A part or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments of the present disclosure.
In addition, functional units in the embodiments of the present disclosure may be integrated into one processing unit, or each of the units may exist alone physically, or two or more units are integrated into one unit. The integrated unit may be implemented in a form of hardware, or may be implemented in a form of a software functional unit.
When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, the integrated unit may be stored in a computer-readable storage medium. Based on such an understanding, the technical solutions of the present disclosure essentially, or the part contributing to the prior art, or all or a part of the technical solutions may be implemented in the form of a software product. The software product is stored in a storage medium and includes several instructions for instructing a computer device (which may be a personal computer, a server, or a network device) to perform all or a part of the steps of the methods described in the embodiments of the present disclosure. The foregoing storage medium includes: any medium that can store program code, such as a USB flash drive, a removable hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc.
The foregoing descriptions are merely specific embodiments of the present disclosure, but are not intended to limit the protection scope of the present disclosure. Any modification or replacement readily figured out by a person skilled in the art within the technical scope disclosed in the present disclosure shall fall within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2014 1 0796168 | Dec 2014 | CN | national |
This application is a continuation of International Application No. PCT/CN2015/096573, filed on Dec. 07, 2015, which claims priority to Chinese Patent Application No. 201410796168.8, filed on Dec. 18, 2014. The disclosures of the aforementioned applications are hereby incorporated by reference in their entireties.
| Number | Name | Date | Kind |
|---|---|---|---|
| 7720377 | Snider | May 2010 | B2 |
| 8260147 | Scandurra | Sep 2012 | B2 |
| 8406623 | Koka | Mar 2013 | B2 |
| 8472808 | Roberts | Jun 2013 | B2 |
| 8606113 | Koka | Dec 2013 | B2 |
| 9479256 | Thonnart | Oct 2016 | B2 |
| 9551836 | Frankel | Jan 2017 | B2 |
| 10117007 | Song | Oct 2018 | B2 |
| 20090274413 | Beausoleil | Nov 2009 | A1 |
| 20100278532 | Scandurra | Nov 2010 | A1 |
| 20110069963 | McLaren | Mar 2011 | A1 |
| 20130294458 | Yamaguchi et al. | Nov 2013 | A1 |
| 20140328208 | Norige et al. | Nov 2014 | A1 |
| 20150139646 | Thonnart | May 2015 | A1 |
| 20150163570 | Zid | Jun 2015 | A1 |
| 20170005735 | Li | Jan 2017 | A1 |
| 20170127161 | Song | May 2017 | A1 |
| 20170289655 | Chen | Oct 2017 | A1 |
| 20180225156 | Lu et al. | Aug 2018 | A1 |
| Number | Date | Country |
|---|---|---|
| 101872039 | Oct 2010 | CN |
| 101917333 | Dec 2010 | CN |
| 10091784 | May 2013 | CN |
| H02143638 | Jun 1990 | JP |
| 101465420 | Nov 2014 | KR |
| 2008067597 | Jun 2008 | WO |
| 2013054497 | Apr 2013 | WO |
| 2014183530 | Nov 2014 | WO |
| Entry |
|---|
| Chen, K., et al., “TWC-based Approach for Improving Communication Reliability in Optical Network-on-chip”, IEEE International Conference of IEEE Region 10 (TENCON 2013), IEEE, Oct. 22, 2013.pp. 1-4. |
| Li, X., et al., “A New Optical Network-on-Chip Architecture for Chip Multiprocessor”, Asia Communications and Photonics Conference (ACP), 2015 OSA, Jan. 1, 2015, 4 pages. |
| Tan, X., et al., “A Hybrid Optoelectronic Networks-on-Chip Architecture”, Journal of Lightwave Technology, vol. 32, No. 5, Mar. 1, 2014, 8 pages. |
| Wang, Z., et al., “Ring Based Optical Network-on-Chip”, Optics Communications, Elsevier, Amsterdam, NL, vol. 285, No. 6, Oct. 14, 2011, pp. 584-596. |
| Ye, Y., et al., “3-D Mesh-Based Optical Network-on-Chip for Multiprocessor System-on-Chip”, IEEE Transactions on Computer-Aided Design of Intergrated Circuits and Systems, vol. 32, No. 4, Apr. 2013, 14 pages. |
| Ye, Y. et al., “3-D Mesh-Based Optical Network-on-Chip for Multiprocessor System-on-Chip,” IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 32, No. 4, Apr. 2013, pp. 584-596. |
| Number | Date | Country | |
|---|---|---|---|
| 20170289655 A1 | Oct 2017 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/CN2015/096573 | Dec 2015 | US |
| Child | 15625025 | US |
